Connecting the Dots: China’s Evolving Standard for Numerical Sub-Range Amendments
Managing global pharmaceutical or chemical portfolios requires navigating China’s notoriously rigid standards on the "same subject matter" requirement for priority and claim amendments. A recent invalidation decision (No. 588094) issued by the China National Intellectual Property Administration (CNIPA) concerning a pharmaceutical patent provides clarity on how numerical ranges and later added experimental data are evaluated in priority claims.1 More specifically, how much literal support is needed for a claim directed towards a “new” numerical sub-range supported by later added working examples?
The Core Dispute: Priority Over specific Ratios and "Added" Data
The patent at issue protected a pharmaceutical composition comprising Edaravone (3-methyl-1-phenyl-2-pyrazolin-5-one) and natural borneol (for treating cerebrovascular diseases), specifically claiming weight ratios from 4:1 to 1:1.2 The challenged patent claimed priority to an earlier application that broadly disclosed weight ranges of 4:1 ~ 1:4 and 2:1 ~ 1:2. Compared to the priority application, the subsequent patent included additional experimental data and new embodiments.
The petitioner challenged the priority claim of the patent, arguing that the later added experimental data and newly described specific weight ratios were not entitled to priority.
The main issue at hand relates to whether certain types of “newly added” content in a patent are entitled to priority to the original filing.
- Newly added experimental data confirming efficacy of certain previously disclosed weight ratios
- Newly added specific weight ratios that falls within a previously disclosed numerical range of weigh ratios but was not specifically disclosed
1. Later added Experimental Data (Priority Sustained for 2:1 and 4:1)
The petitioner argued that because the subsequent Chinese application added several new efficacy tables and experimental models that were completely absent from the priority document, the subsequent invention could not be considered the "same subject matter" as that which was in the priority application.
The CNIPA panel ruled that adding experimental data in a subsequent application does not automatically destroy its priority claim. If a person skilled in the art can confirm (based on the original priority disclosure) that the technical field, the underlying problem solved, and the expected technical effects remain identical, the newly added data is viewed merely as a further confirmation and refinement of the original effects. Because the priority document explicitly disclosed the 2:1 and 4:1 ratios and verified basic efficacy, the priority for those specific embodiments was upheld.
2. The Strictness of Support within Numerical Ranges (Priority Denied for 1:1)
Conversely, while the priority document disclosed a broad preferred range (4:1 ~ 1:4 and 2:1 ~ 1:2) that technically encompassed a 1:1 ratio, it never explicitly highlighted "1:1" as a distinct, standalone embodiment.
CNIPA strictly adheres to the "direct and unambiguous derivation" standard for numerical features. Because "1:1" was not expressly disclosed or directly determinable from the priority text, the panel denied priority for the 1:1 ratio claim. (Note: While the 1:1 claim ultimately survived the obviousness challenge based on other merits, losing the priority date exposed it to a year's worth of potential prior art).
The table below breaks down the exact differences that led the panel to uphold priority for the 4:1 and 2:1 ratios, while denying it for the 1:1 ratio.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Eagle Thoughts
This case clarifies the CNIPA’s approach to two separate aspects of sufficiency: data support (“enablement”) and literal support (“basis”).
Consistent with prior cases, rules, regulations, and guidelines, China is very strict when it comes to giving “credit” (e.g., basis/support or a priority claim) for later submitted data and/or ideas. Any subject matter or protection scope must be clearly described or could be undoubtedly obtained by skill in the art based on the provided information at the time of filing to be considered “supported” as of the earlier date. We saw this, for example, in China’s recent decisions on post-filing supplemental data.
This case further clarifies that numerical ranges alone do not provide sufficient priority for specific (previously undisclosed) intermediate values. However, disclosure of the specific endpoints may provide support for a “new” sub-range encompassed by those two previously disclosed endpoints.
So what drafting strategy is appropriate for an applicant who has identified working numerical ranges but has not yet “fleshed out” all the working examples?
File Early and Draft Many Layers of Ranges and “Dots” within the Ranges:
Before this case, we would advise applicants to draft the specification with many different layers (e.g., various ranges, sub-layers of ranges), with plans to aggressively generate experimental data to include in the subsequent PCT or national application. This is because the CNIPA has been traditionally strict with regards to literal support (basis) and would require literal support of a particular sub-range or sub-layer. This still holds true, especially for the unpredictable arts like chemical small molecules.
However, as we have seen from this case, the CNIPA seems to treat numerical ranges and ratios a bit differently because they are considered more routine and predictable. In these types of situations, specific numbers can provide literal support (basis) for “new” ranges having such specific numbers as endpoints. This expands the possible sub-scopes that are available to be supported by subsequent working examples.
This “relaxed” standard tilts the timeline for filing that first priority application a bit earlier, particularly for applicants with inventions involving ratios and/or numerical ranges. In the initial specification, list out not only layers of ranges, but also explicitly “dot” the specification with target numbers within a range. Later, when the applicant provides follow-on working examples, the CNIPA should recognize literal support for “new” sub-ranges encompassed by such “dotted” numerical endpoints. Having this type of literal support will strengthen a patent a lot, allowing the applicant to amend to a “new” sub-scopes that cover working examples having the strongest data.
The Bigger Picture:
Any drug protection strategy in China must include a China-focused patent strategy that works in conjunction with broader considerations, such as the standards for priority and support in other jurisdictions, the prior art effects of “secret prior art”, patent term adjustment, as well as IP-adjacent exclusivity considerations such as patent linkage, patent term extension, data exclusivity, and more. Still, knowing China’s approaches numerical ranges helps practitioners weigh the pros and cons of filing earlier, especially with regards to being able to claim priority for “new” numerical sub-ranges surrounding later added working examples.
If you would like to have more information on this matter or would like to have our advice, please feel free to contact us at [email protected].
This article is for general informational purposes only and should not be considered legal advice or a legal opinion on a specific set of facts.
Top 10 Cases of Patent Re-examination Board (CNIPA) 2025.
↩︎ZL 200980100527.9, filed on 3 Mar 2009, claiming priority of 4 Mar 2008. Filed by Simcere Pharmaceutical R&D Co Ltd
↩︎
Claim 1: 一种药物组合物，包括3-甲基-1-苯基-2-吡唑啉-5-酮和天然冰片，其特征在于，所述3-甲基-1-苯基-2-吡唑啉-5-酮和天然冰片的重量比为4∶1～1∶1.
A pharmaceutical composition comprising 3-methyl-1-phenyl-2-pyrazolin-5-one and natural borneol, wherein the weight ratio of 3-methyl-1-phenyl-2-pyrazolin-5-one to borneol is from 4:1 to 1:1.
